Learning Vision-based Reactive Policies for Obstacle Avoidance
This addresses the problem of enabling robots to avoid obstacles using vision, which is incremental as it builds on existing work by integrating perception and motion in a unified way.
The paper tackles vision-based obstacle avoidance for robotic manipulators by proposing a unified framework that connects perception and motion, learning reactive policies that achieve high success rates in goal-reaching tasks with single and multiple obstacles.
In this paper, we address the problem of vision-based obstacle avoidance for robotic manipulators. This topic poses challenges for both perception and motion generation. While most work in the field aims at improving one of those aspects, we provide a unified framework for approaching this problem. The main goal of this framework is to connect perception and motion by identifying the relationship between the visual input and the corresponding motion representation. To this end, we propose a method for learning reactive obstacle avoidance policies. We evaluate our method on goal-reaching tasks for single and multiple obstacles scenarios. We show the ability of the proposed method to efficiently learn stable obstacle avoidance strategies at a high success rate, while maintaining closed-loop responsiveness required for critical applications like human-robot interaction.